The CirCor DigiScope Phonocardiogram Dataset

Abstract

A total number of 5282 heart sound recordings were collected from the main four auscultation locations of 1568 patients, aged between 0.1 and 356.1 months (mean ± STD = 73.4 ± 50.3 months), with a duration between 4.8 to 80.4 seconds (mean ± STD = 22.9 ± 7.4 s), summing up to more than 312 hours of recording. For the first time, each cardiac murmur in the dataset has been annotated in detail, in terms of timing, shape, pitch, grading, quality, and location, by using a semi-supervised annotation scheme. The annotations were performed by voting between three state-of-the-art machine-based algorithms. Two independent experts later studied the consensus and mismatches between the algorithms beat-by-beat, and performed a manual annotation whenever the algorithms had disagreed. To date, the dataset is the largest pediatric heart sound dataset and it paves the way to a deeper research on the topic of auscultation-based health recommendation systems. The dataset is available online for public research on PhysioNet.

Background

Normal heart sounds are generated from vibrations of the cardiac valves as they open and close during the cardiac cycle. The anatomical position of heart valves relative to the chest wall dictates the optimal auscultation position. As such, for clinical auscultations of each heart valve, the stethoscope is ideally placed at specific locations: 

• Aortic valve: second intercostal space, right sternal border
• Pulmonic valve: second intercostal space, left sternal border
• Tricuspid valve: left lower sternal border
• Mitral valve: fifth intercostal space, midclavicular line (cardiac apex)

Blood flowing through these structures creates audible sounds, which are more audible as more turbulent the flow is [1]. The first heart sound (S1) is produced by vibrations of the mitral and tricuspid valves as they close in at the beginning of the systole. S1 is audible at the chest wall and is formed by two components – the mitral and tricuspid [1]. Although the mitral component of S1 is louder and occurs sooner, under physiological resting conditions, both components occur closely enough, making it hard to distinguish them [2]. The second heart sound (S2) is produced by the closure of the aortic and pulmonic valves, at the beginning of the diastole. Similarly to the S1, it is also formed by two components, with the aortic component being louder and occurring sooner than the pulmonic component, due to the pressures in the aorta being higher than in the pulmonary artery. In contrast and unlike S1, under normal conditions the closure sound of the aortic and pulmonic valves can be discernible, due to an increase in venous return during inspiration, which slightly delays the pressure increase in the pulmonary artery and consequently the pulmonic valve closure [3].

Methods

The dataset was collected as part of two mass screening campaigns conducted in Northeast Brazil between July-August 2014 and June-July 2015 [4]. The data collection was approved by the 5192-Complexo Hospitalar HUOC/PROCAPE institutional review board, under the request of the Real Hospital Portugues de Beneficencia em Pernambuco. The target population included all participants presenting voluntarily for screening within the study period. Patients younger than 21 years of age with a parental signed consent form (where appropriate) were included. A total of 2061 participants attended the 2014 and 2015 “Caravana do Coração” (Portuguese for “Caravan of the Heart”) campaigns, with 493 participants being excluded for not meeting the eligibility criteria. All participants completed a socio-demographic questionnaire and subsequently underwent a clinical examination (anamnesis and physical examination), a nursing assessment (physiological measurements), and cardiac investigations (chest radiography, electrocardiogram, and echocardiogram). Data quality assessment was performed, and all entries were screened for incorrectly entered or measured values, inconsistent data or presence of outliers, and deleted as appropriate. The resulting entries were then compiled in a spreadsheet to reflect the socio-demographic and clinical variables used in our dataset. 

Subsequently, an electronic auscultation was performed and audio samples from four typical auscultation points were collected. All samples were collected by the same operator for the duration of the screening, in a real clinical setting. Two independent cardiac physiologists individually assessed the resulting phonocardiogram (PCG) audio files for signal quality. As a result, 119 participants did not meet the required signal quality standards, i.e. these patient recordings do not lead into a reliable murmur characterization and description.

The acquired audio samples were automatically segmented using the three algorithms proposed in [5], [6] and [7]. These algorithms were only used to detect and identify the fundamental heart sounds (S1 and S2 sounds) and their corresponding boundaries. Two independent cardiac physiologists inspected the resulting algorithms’ outputs on mutually exclusive data. In other words, they did not over-read each other’s work, therefore providing independent annotations. Each expert analyzed the automated annotations and whenever the annotator disagreed with the suggested automatic annotations, a manual annotation was required. In such cases, the annotator was instructed to annotate at least five complete heart cycles. Segmentation labels were retained for sections of heart sound recordings that were considered of high quality and representative by the cardiac physiologists. The remainder of the signal may include both low and high quality data. In this way, the users of the dataset may choose to use (or not to use) the suggested time windows, where the signal quality was manually inspected, and the automated labels were validated.

This methodology was applied to both the 2014 and 2015 screening campaigns, resulting in two independent folders (CC2014, CC2015). A cardiac physiologist manually classified and characterized murmur events blindly, and independent of other clinical notes. The cardiac physiologist inspected by listening and visualizing simultaneously each non-filtered heart sound signal, through the Audacity application. Note that no segmentation data is used by the expert in order to detect and characterize murmur events.  The sounds were recorded in an ambulatory environment. Different noisy sources have been observed in our dataset, including stethoscope rubbing noise, speaking, crying, or laughing sounds in the background. On the other hand, the proposed dataset is a representative sample of real case environments where automatic machine-based auscultation systems should operate. Further details regarding the methodology can be found in [1].

Data Description

The audio and the corresponding segmentation annotation file names are in ABCDE_XY.wav and ABCDE_XY..txt format, respectively. Here ABCDE is a numeric patient identifier and XY is one of the following codes corresponding to the auscultation location where the PCG was collected on the body surface: PV corresponds to the Pulmonary point; TV corresponds to the Tricuspid point; AV corresponds to the Aortic point, MV corresponds to the Mitral point, and finally PhC for any other auscultation. location. If more than one recording exists per auscultation location, an integer index proceeds the auscultation location code in the file name, i.e, ABCDE_nXY.wav, where n is an integer. Furthermore, each audio file has its own corresponding  annotation segmentation file.

The segmentation annotation file is composed of three distinct columns: the first column corresponds to the time instant where the wave was detected for the first time; the second column corresponds to the time instant where the wave was detected for the last time; the third column corresponds to an identifier that uniquely identifies the detected wave, namely:

• The S1 wave is identified by the integer 1.
• The systolic period is identified by the integer 2.
• The S2 wave is identified by the integer 3.
• The diastolic period is identified by the integer 4.
• Not annotated segments of the signal are identified by the integer 0.

All the collected heart sound records were also screened for presence of murmurs at each auscultation location. Each murmur was classified  according to its timing (early-, mid-, and late- systolic/diastolic) [8], shape (crescendo, decrescendo, diamond, plateau), pitch (high, medium, low), quality (blowing, harsh, musical) [8], and grade (according to Levine’s scale [9]). The aforementioned information is displayed and described on a patient basis on both the training_data.csv and testing_data.csv files respectively (see Section Usage Notes for more details).

Usage Notes

The dataset is organized in two sets: training and testing sets. Currently, only 30% of the gathered dataset is publicly released (randomly selected through stratified random sampling over the following classes: normal patient, abnormal patient and unsure) for training purposes. An additional 40% of the data is going to be released in the near future, before a forthcoming international data challenge starts.

The remaining 30%, which is currently held as a private repository for test purposes of a forthcoming international data challenge, will be released in future versions of the same PhysioNet project, after the corresponding data challenge.

For each of the train/test subsets of the data, there is a corresponding CSV file, namely training_data.csv and testing_data.csv (the latter will be released after the full release of the challenge subset). These files have the same format (i.e., the same variables), except that they describe different and exclusive sets of the studied patients. In these files, murmur waves are described and categorized per patient and following the same terminology used by physicians, namely its timing, shape, pitch, grading and quality. In addition, the auscultation locations where the murmur is present as well as the auscultation location where the murmur is detected more intensively are also reported in these files.

The first column is named Patient_ID, the entries on this column are identifiers used to uniquely determine each patient on the database. These identifiers are also been used to link heart sound signals and annotation files to the murmur characterization data in the *.csv files.

The second column is titled Murmurs, each entry on this column is an integer between 1 and 3, with the following meaning:

• 1: Murmur waves were detected in at least one heart sound recording.

• 2: Murmur waves were not detected in any heart sound recording.

• 3: Inconclusive.

When murmurs are not present, the entries in the remaining columns are NA (Not Applicable). Instead, when the annotator could not make any decision (Inconclusive), the entries in the remaining columns are 0.

The third column is titled Murmurs_Location. The entries on this column are string type, each string is a concatenation of acronyms. Each acronym represents an auscultation spot where at least one murmur wave has been observed. If more than one spot needs to be reported, the spots are listed and separated by the plus (+) sign.

The acronyms used to identify the auscultation locations names are: PV (the Pulmonary location); TV (Tricuspid point); AV (Aortic point), MV (Mitral spot), and PhC for any other auscultation location.

The fourth column is titled Most_Audible. The entries on this column are string type, each string is a single acronym that identifies the auscultation point where murmur waves were observed more intensively. The acronyms used are the same as in Murmur_Location.

The fifth through ninth columns are exclusively used to describe and categorize murmurs that are located in the systolic period.

The fifth column is titled Murmur_Systolic_Timing. This column describes the location of the murmur wave in the systolic period. The possible entries for this variable are:

• Early-systolic: a murmur has been observed at the beginning of the systolic period.

• Holosystolic: a murmur has been observed over the whole systolic period.

• Late-systolic: a murmur has been observed at the ending of the systolic period.

• Midsystolic: a murmur has been observed at the middle of the systolic period.

The sixth column is titled Murmur_Systolic_Shape. This column describes the shape of the murmur wave that has been observed in the systolic period. The shape of a murmur can be viewed as a function of murmur intensity over time. The possible entries for this variable are:

• Crescendo: the amplitude of the murmur wave increases over time.

• Decrescendo: the amplitude of the murmur wave decreases over time.

• Diamond: the amplitude of the murmur wave first increases for an amount of time but then decreases for the rest of the time period.

• Plateau: the amplitude of the murmur wave stays approximately constant over the whole period.

The seventh column is titled Murmur_Systolic_Pitch. This column describes the murmur's pitch feature from waves observed in the systolic period. The pitch variables are related to the pressure gradient felt in the heart chambers. In general the higher the pitch is, the higher is the pressure gradient felt in the corresponding heart chamber. For example, in an aortic stenosis, a large pressure gradient is felt between the left ventricle and the aorta artery. As a result, murmurs generated by an aortic stenosis have in general a high pitch. The possible entries for this variable are High, Medium, and Low.

The eight column is titled Murmur_Systolic_Grading. This column describes the murmur's grade feature from waves observed in the systolic period. It is highly correlated with the severity of the murmur. The higher the grading, the worse is the patient prognostic and outcome.

Since not all patients have auscultation sounds recorded from all the four main auscultation locations, the strategy adopted to provide grading annotations is described as follow:

• Grade I/IV: if barely audible and not heard/present or not recorded in all auscultation locations;

• Grade II/IV: if soft but easily heard in all auscultation locations;

• Grade III/IV: if moderately loud or loud.

Accordingly, the grade annotations can diverge from the original definition of murmur grading, when applied to cases for which not all the auscultation locations are available. In such cases, murmurs were classified by default as grade I/VI. Moreover, the cases classified as grade III/VI, actually include murmurs that could potentially be of grade III/VI or higher, since discrimination among grades III/VI, IV/VI, V/VI, and VI/VI is associated with palpable murmurs, also known as thrills [9], which can only be assessed via physical in-person examination.

The nine column is titled Murmur_Systolic_Quality. This column describes the murmur's quality feature from waves observed in the systolic period. It relates to the presence of harmonics and the overtones. The possible entries for this variable are Blowing, Harsh, Musical.

The tenth through fourteenth columns describe the murmurs that have been observed in the diastolic period. The same set of features considered for waves observed in the systolic period are considered: timing, shape, pitch, grading, and quality.

The tenth column is titled Murmur_Diastolic_Timing. This column describes the location of the murmur wave in the diastolic period. The possible entries for this variable are:

• Early-diastolic: a murmur has been observed at the beginning of the diastolic period.

• Mid-diastolic: a murmur has been observed at the middle of the diastolic period.

The eleventh column is titled Murmur_Diastolic_Shape. This column describes the shape of the murmur wave that has been observed in the diastolic period. In the gathered database, only murmurs with Decrescendo and Plateau shapes have been observed in the diastolic period.

The twelfth column is titled Murmur_Diastolic_Grading. This column describes the murmur's grade feature from waves observed in the diastolic period. In contrast to systolic murmurs, diastolic murmurs do not follow the Levine grading scale [9]. Instead murmurs are graded from I to IV (instead of I to VI).

• Grade I/IV: if barely audible and not heard/present or not recorded in all auscultation locations;
• Grade II/IV: if soft but easily heard in all auscultation locations;
• Grade III/IV: if moderately loud or loud.

On the other hand, IV/IV are associated with palpable murmurs, also known as thrills [9], which can only be assessed via physical in-person examination. In the current database Grades III/IV and IV/IV are merged together.

The thirteenth column is titled Murmur_Diastolic_Pitch. This column describes the murmur's pitch feature from waves observed in the diastolic period. The possible entries for this variable are Medium, Low and High.

The fourteenth column is titled Murmur_Diastolic_Quality. This column describes the murmur's quality feature from waves observed in the diastolic period. In our databases, only murmurs with Blowing and Harsh qualities have been observed in the diastolic period.

The fifteenth column is titled Gender. The entries on this column are Male or Female.

The sixteenth column is titled Age_Group, corresponding to the age category (child, infant, neonate, adolescent and young adult) of the patient following the National Institute of Child Health and Human Development (NICHD) pediatric terminology [10].

The seventeenth column is titled Height, each entry provides the patient height in centimeters (cm).

The eighteenth column is titled Weight, which corresponds to the patient's weight in kilograms.

The nineteenth column is titled isPregnant and explicitly identifies the subjects that were pregnant at the time of the screening campaign.

Since some patients attended the two screening campaigns but using a different identifier, the twentieth column titled PatientID_in_otherCC provides the second identifier used for the patient. Furthermore, in order to remove any data bias in our dataset, data from the same patient (regardless of the screening campaign) is either in the training or the testing set (the latter not provided in the current release).

The last column is titled Campaign. The entries on this column are of string type. Each string is a single acronym that identifies the screening campaign that the patient attended. The acronyms used are the following:

• CC2014: the 2014 screening campaign;
• CC2015: the 2015 screening campaign.

Note that the two screening campaigns have a one year time gap and taking into account the patient's age, an open question that requires further study is whether or not any long-term dependencies can be found between the data acquired from the same subjects between the two campaigns. We let this discussion open to the scientific community for further evaluation.

Any research/publication based on this database is requested to cite the CirCor DigiScope Phonocardiogram Dataset project on PhysioNet and its corresponding article [4].

Release Notes

Initial release.

Acknowledgements

This work is a result of the Project DigiScope2 (POCI-01-0145-FEDER-029200 - PTDC/CCI-COM/29200/2017), funded by Fundo Europeu de Desenvolvimento Regional (FEDER), through Programa Operacional Competitividade e Internacionalização (POCI), and by national funds, through Fundação para a Ciência e Tecnologia (FCT).

Conflicts of Interest

The authors declare that there are no conflicts of interest.

References

1. P. Libby, R. Bonow, D. Mann, and D. Zipes, Braunwald’s Heart Disease: A Textbook of Cardiovascular Medicine. 8th edition. Elsevier Science, 2007.
2. J. Soler-Soler and E. Galve, “Worldwide perspective of valve disease,” Heart, vol. 83, no. 6, pp. 721–725, 2000. [Online]. Available: https://heart.bmj.com/content/83/6/721
3. T. A. Dornbush S. (2019) Physiology, heart sounds. [Online]. Available: In:StatPearls[Internet].TreasureIsland(FL):StatPearlsPublishing, https://www.ncbi.nlm.nih.gov/books/NBK541010/
4. J. H. Oliveira, F. Renna, P. Costa, D. Nogueira, C. Oliveira, C. Ferreira, A. Jorge, S. Mattos, T. Hatem, T. Tavares, A. Elola, A. Rad, R. Sameni, G. D. Clifford, & M. T. Coimbra (2021). The CirCor DigiScope Dataset: From Murmur Detection to Murmur Classification. IEEE Journal of Biomedical and Health Informatics, https://doi.org/10.1109/JBHI.2021.3137048
5. C. Liu, D. Springer, Q. Li, and et. al., “An open access database for the evaluation of heart sound algorithms,” Physiological Measurement, vol. 37, no. 12, p. 2181, 2016
6. J. Oliveira, F. Renna, and M. T. Coimbra, “Adaptive sojourn time HSMM for heart sound segmentation,” IEEE J. Biomed. Health Informatics, vol. 23, no. 2, pp. 642–649, 2019.
7. F. Renna, J. H. Oliveira, and M. T. Coimbra, “Deep convolutional neural networks for heart sound segmentation,” IEEE Journal of Biomedical and Health Informatics, vol. 23, no. 6, pp. 2435–2445, 2019.
8. S. J. Owen and K. Wong, “Cardiac auscultation via simulation: a survey of the approach of uk medical schools,” BMC research notes, vol. 8, pp. 427–427, Sep 2015, 26358413 [pmid]. [Online]. Available: https://pubmed.ncbi.nlm.nih.gov/26358413
9. A. Freeman and S. Levine, “The clinical significance of the systolic murmur. a study of 1000 consecutive “non-cardiac” cases,” Ann Intern Med, vol. 6, p. 1371–1385, 1933.
10. K. Williams, D. Thomson, I. Seto, D. Contopoulos-Ioannidis et al., “Standard 6: Age groups for pediatric trials,” Pediatrics, vol. 129 Suppl 3, pp. S153–60, 06 2012